Gene prioritization for livestock diseases by data integration

Abstract

Identifying causal genes that underlie complex traits such as susceptibility to disease is a primary aim of genetic and biomedical studies. Genetic mapping of quantitative trait loci (QTL) and gene expression profiling based on high-throughput technologies are common first approaches toward identifying associations between genes and traits; however, it is often difficult to assess whether the biological function of a putative candidate gene is consistent with a particular phenotype. Here, we have implemented a network-based disease gene prioritization approach for ranking genes associated with quantitative traits and diseases in livestock species. The approach uses ortholog mapping and integrates information on disease or trait phenotypes, gene-associated phenotypes, and protein-protein interactions. It was used for ranking all known genes present in the cattle genome for their potential roles in bovine mastitis. Gene-associated phenome profile and transcriptome profile in response to Escherichia coli infection in the mammary gland were integrated to make a global inference of bovine genes involved in mastitis. The top ranked genes were highly enriched for pathways and biological processes underlying inflammation and immune responses, which supports the validity of our approach for identifying genes that are relevant to animal health and disease. These gene-associated phenotypes were used for a local prioritization of candidate genes located in a QTL affecting the susceptibility to mastitis. Our study provides a general framework for prioritizing genes associated with various complex traits in different species. To our knowledge this is the first time that gene expression, ortholog mapping, protein interactions, and biomedical text data have been integrated systematically for ranking candidate genes in any livestock species.

Early differences in islets from prediabetic NOD mice: combined microarray and proteomic analysis

“…The networks were visualised in Cytoscape [22]. Ranking of the differentially expressed mouse proteins based on their assignment to type 1 diabetes relevant protein complexes was performed as described previously [23]. PANTHER was used to classify genes and proteins by biological processes [24].…”

Exploring the genetic architecture and improving genomic prediction accuracy for mastitis and milk production traits in dairy cattle by mapping variants to hepatic transcriptomic regions responsive to intra-mammary infection

“…Gram-negative Escherichia coli ( E. coli ) is a common mastitis-causing bacteria [22], and the lipopolysaccharides (LPS) released by E. coli induce acute inflammatory responses [23]. Genes with expression levels that are significantly affected during the early stage of infection have also been suggested to be involved in overall metabolism [19, 23–26]. Moreover, it is well established that mastitis is unfavorably correlated with milk production traits [25].…”

A random set scoring model for prioritization of disease candidate genes using protein complexes and data-mining of GeneRIF, OMIM and PubMed records

“…PubMed may capture the phenotype-genotype relationships more comprehensively than OMIM does. This has important implications for disease gene prioritization in livestock species [ 26 ], for which only a limited number of phenotypic descriptions have accumulated in disease databases (e.g., OMIA), although many more gene-associated phenotypes may exist in literature databases such as PubMed or GeneRIF. The performance evaluation using the set of known causal genes from OMIM could be potentially overestimated even if the associations between a test gene and diseases (including test disease) were excluded from validation study.…”